Throughout the brain, specialized systems carry out different but complementary functions, sometimes independently but often in cooperation. However, we do not understand how their activity is dynamically coordinated, and dysregulation of this is associated with many mental health conditions. Neuronal oscillations, which are detectable in local field potentials (LFPs) at various frequencies, are a promising target for this coordination. Gamma oscillations (40-100 Hz) in particular have been singled out since they enhance stimulus responses, facilitate interactions between brain regions, and are expressed ubiquitously across cortical and subcortical regions. Indeed, gamma oscillations occur in the basolateral nucleus of the amygdala (BL), an important regulator of emotional behaviors. BL gamma oscillations are enhanced during periods of heightened vigilance during a foraging task, following emotionally salient experiences, and upon presentation of socially-relevant stimuli. The variety of circumstances that engage it make it a promising target for interventions affecting emotional behaviors in general. However, technical challenges abound because gamma manifests as brief intermittent oscillatory bursts, layered atop numerous ongoing activities in other frequency bands. This precludes manipulating gamma exclusively with traditional pharmacological, optogenetic, or chemogenetic approaches, since these have substantial effects on ongoing non-gamma activities, and are delivered irrespective of whether gamma bursts are present or absent. To overcome this, a closed-loop algorithm was developed that monitors the LFP in real-time for gamma oscillations and delivers precisely timed optogenetic stimulation capable of enhancing or suppressing gamma strength on a cycle-by-cycle basis. While this improves upon the status quo,, further refinement is needed. Aim 1 of this proposal seeks to clarify how the gamma modulation technique operates via biophysically detailed modeling of the local circuits in the BL that generate gamma, the effects of optogenetic stimulation, and the closed-loop algorithm. Aim 2 designs better signal processing routines for detecting and parameterizing gamma in real-time. Aim 3 develops an approach to create customized biophysical models that reproduce the properties of gamma observed in individual subjects, which when combined with the results of Aims 1 and 2 should allow for optimized control over gamma oscillations in individual subjects. RELEVANCE (See instructions): Gamma oscillations occur in the basolateral amygdala, a brain region implicated in emotional regulation. By developing improved methods to manipulate these oscillations, we hope to better understand their function and improve our ability to control emotional states and behaviors.